Nuclear fuels: uranium, MOX, HALEU, thorium

Behind arguments about nuclear power stations sits a quieter, more strategic issue: the fuels themselves-how they are manufactured, where they are sourced, and which nations are positioned to control supply over the coming decades.

Energy density that defies common sense

At the level of atoms, nuclear fuels deliver extraordinary energy. A single fission releases roughly 200 MeV (million electronvolts), which works out at close to 80 million megajoules per kilogram of fuel.

By contrast, coal provides about 24 megajoules per kilogram. Measured by mass, fission is therefore on the order of 10 million times more energy-dense than burning coal.

• 1 kg of uranium fuel: potentially enough energy to keep a city’s lights on for days
  
• 1 kg of coal: consumed in minutes in a power station boiler

Even among fissile isotopes, the details influence reactor choices:

Isotope	Energy per fission (MeV)	Average neutrons released	Typical use
Uranium‑235	~193	~2.45	Conventional thermal reactors
Plutonium‑239	~199	~2.9	Fast reactors, MOX
Uranium‑233	~191	~2.5	Thorium‑based cycles

That higher neutron yield from plutonium is a key reason it is so appealing for fast breeder reactors, which can be designed to create more fuel than they consume.

The fuels quietly powering today’s reactors

Most operating reactors rely on a small family of fuels and isotopes. Each brings its own engineering trade-offs-and its own geopolitical importance.

Uranium: the workhorse the grid still depends on

Although natural uranium is widely found, only a tiny fraction is readily fissionable. Just 0.72% is uranium‑235 (U‑235), the isotope that sustains the chain reaction; most of the remainder is uranium‑238 (U‑238), which largely plays a supporting role in today’s mainstream reactor types.

To make uranium suitable for most reactors, it is enriched so the U‑235 proportion increases to roughly 3–5%. This product-low enriched uranium (LEU)-is the standard fuel for the bulk of the world’s pressurised water and boiling water reactors.

LEU is the understated foundation of global nuclear generation: a mature, standardised fuel supported by a complete industrial chain from the mine through to the spent-fuel pool.

LEU’s defining strength is reliability. Regulators and operators understand its behaviour, and a supply chain dominated by a limited number of major firms has been refined over many decades.

An often-overlooked step sits between mining and enrichment: uranium is typically converted into chemical forms that can be processed efficiently (including conversion to gaseous compounds used in centrifuges), then fabricated into precisely engineered fuel assemblies. Each stage adds cost, quality requirements and-increasingly-strategic leverage for the countries that host the facilities.

MOX: turning “waste” plutonium back into fuel

Mixed oxide fuel (MOX) uses plutonium recovered from spent fuel and blends it with depleted uranium. In practical terms, material once treated as a disposal burden is repurposed as an energy resource.

In a closed fuel cycle, using MOX can reduce the need for freshly mined natural uranium by around 20%. France is frequently cited as the reference case: its industrial recycling approach routinely loads MOX assemblies into part of its reactor fleet.

This route can be attractive to countries anxious about long-term uranium prices, but it brings demanding chemistry, stringent safeguards, and higher initial costs.

HALEU: the coming fuel for small modular reactors (SMRs)

HALEU means High Assay Low Enriched Uranium. It sits between conventional LEU and weapons-grade material, with enrichment in the range of 5–20% U‑235.

That band is especially useful for many small modular reactors (SMRs) and several Generation IV concepts. With a higher share of fissile atoms, designers can shrink the core and extend operation, reducing the frequency of refuelling outages.

HALEU is attractive because it supports longer fuel cycles, smaller reactor cores and fewer shutdowns-precisely the gains many next‑generation designs are aiming for.

The constraint is availability. Only a limited number of facilities can manufacture HALEU at meaningful scale, and a significant portion of that capability is linked to Russia, which has triggered concern in many Western capitals that are simultaneously trying to accelerate SMR programmes.

TRISO: fuel engineered not to melt

TRISO fuel-short for tristructural‑isotropic-is not a simple pellet. It consists of countless tiny uranium “kernels”, each sealed inside multiple protective layers of ceramic and carbon, like a microscopic onion.

These particles can withstand temperatures above 1,600°C while keeping fission products contained. This makes TRISO well suited to high‑temperature gas‑cooled reactors, where designers target “walk‑away safety”-severe incidents are intended to be unable to easily compromise the fuel itself.

The downside is manufacturing difficulty and cost: producing millions to billions of near-flawless particles is technically demanding, and that complexity is reflected in the price.

Thorium: the slow-burn challenger

Thorium‑232 is not fissile on its own. Inside a reactor it can absorb a neutron and, through subsequent transformations, become uranium‑233 (U‑233)-a fissile isotope with broadly similar reactor-relevant characteristics to U‑235.

India and China, both with substantial thorium resources, treat this as a long-horizon strategic option. They are investing heavily in molten‑salt reactors and other designs that would rely on thorium-centred fuel cycles.

Thorium is not a quick fix; it is a slower-moving alternative that could materially change fuel security in the latter half of the century.

Advocates point to its relative abundance and the potential to reduce certain long-lived waste components. Sceptics counter that much of the end-to-end industrial system-fuel fabrication, handling and reprocessing-would need to be developed at scale from near zero.

Open, closed and alternative cycles: what happens to spent fuel?

The choice of fuel is only half the story. What countries do with spent fuel is central to cost, waste, and proliferation risk.

Open cycle: use once, store forever

Many states-including the United States-operate an open cycle. After use, fuel is cooled in pools and then transferred into dry casks for long-term storage, without chemical reprocessing.

A gigawatt-scale pressurised water reactor operating for one year produces about 28.8 tonnes of highly radioactive spent fuel, in addition to large quantities of mining residues upstream.

This approach minimises industrial complexity today, but it leaves future societies responsible for guarding long-lived materials for centuries and beyond.

Closed cycle: recycle and shrink the waste footprint

In closed cycles, practised by France, Russia and a small number of others, spent fuel is reprocessed to separate uranium and plutonium. The plutonium is reused in MOX, while recovered uranium can be re‑enriched or held for possible future fast reactors.

Recycling can reduce the final volume of high-level waste by around a factor of four. However, the remaining waste is more heat-generating in the near term, and reprocessing must be conducted under strict safeguards to manage proliferation risks.

Thorium cycle: fewer long‑lived nasties

A major selling point of thorium pathways is the lower production of minor actinides-long-lived elements that persist for hundreds of thousands of years and dominate the long-term radiotoxicity profile of conventional nuclear waste.

There is also a proliferation-related twist: U‑233 bred from thorium commonly contains traces of uranium‑232, which produces intense gamma radiation. That contamination makes any hypothetical military diversion materially more difficult, an attribute welcomed by many non-proliferation specialists.

Fuel-cycle choices also shape transport and security burdens. Moving fresh LEU, MOX, HALEU or specialised TRISO products involves different packaging standards, guard requirements and regulatory scrutiny. As reactor fleets diversify, so do the logistical and safeguards demands-particularly if many smaller sites replace a smaller number of large stations.

Reserves and geopolitics: who owns the atoms?

Uneven uranium, better‑spread thorium

Recoverable uranium resources are estimated at roughly 7.9 million tonnes. Annual demand is about 69,000 tonnes, and it could more than double by 2040 if the widely discussed nuclear expansion actually occurs.

Australia holds the largest uranium reserves, with Kazakhstan and Canada also prominent. Yet Kazakhstan leads production, supplying over 40% of global mined output via its state-backed company Kazatomprom.

Control over mining and enrichment is starting to carry the same political sensitivity that gas pipelines did in the 2000s.

Thorium resources are estimated at about 6.3 million tonnes. It is three to four times more abundant in the Earth’s crust and more evenly distributed. India, the United States and Australia all have significant deposits, which could reduce the chance of any single country dominating supply if thorium reactors ever scale.

Who holds the key pieces of the fuel market?

Mining and enrichment as strategic choke points

On the mining side, Kazatomprom, Cameco (Canada) and Orano (France) form a small group of heavyweights. In enrichment, the concentration is even sharper.

Rosatom and its subsidiary Tenex are widely cited as controlling around 40–50% of global enrichment capacity. The European consortium Urenco holds roughly 30%, and Orano accounts for a smaller yet still meaningful portion.

Replacing Russian enrichment capacity would be neither quick nor simple: new centrifuge plants take years to build and qualify.

Fabrication and advanced fuels

For fuel fabrication, suppliers such as Westinghouse and Framatome manufacture LEU fuel assemblies for reactor fleets across Europe and Asia. For MOX, France’s Melox facility (Orano) remains one of the few industrial-scale plants in operation.

In the United States, companies including Centrus and BWXT are pushing to establish dependable HALEU production for SMRs and advanced reactors. Without a robust HALEU pipeline, many highly publicised “reactors of the future” risk delays for a very basic reason: the right fuel will not be available when needed.

Beyond fission: how fusion frames the debate

Investors and policymakers often ask whether fusion will make today’s fuel questions irrelevant. For now, that remains speculative.

Fusion relies on hydrogen isotopes-deuterium and tritium-rather than uranium or plutonium. The leading deuterium–tritium reaction releases about 17.6 MeV per event and is often described as delivering roughly four times the energy per kilogram compared with fission fuels.

However, tritium supply is a major challenge. It must be bred from lithium in specialised blankets around the plasma, and no commercial-scale system has yet demonstrated that it can reliably “close” that breeding loop.

ITER, the large experimental fusion project in southern France, is intended to prove that fusion can generate more energy than it consumes. Even under optimistic assumptions, commercial fusion before the 2040s would be hard to achieve-meaning fission fuels are set to remain central for decades.

Key concepts readers often ask about

Actinides, toxicity and time scales

A common question is what makes nuclear waste hazardous for so long. In the first centuries, much of the danger comes from fission products, many of which decay substantially over that timeframe. Over the very long term, the dominant concern is actinides-heavy elements such as plutonium, americium and curium.

Closed cycles and prospective fast reactors aim to burn or transmute a larger share of these actinides, shortening the period during which waste requires extreme isolation. Thorium-based routes may help by producing fewer of these elements in the first place.

What an HALEU‑fuelled grid could look like (SMRs, HALEU and industrial heat)

Some analyses envisage dozens or even hundreds of HALEU‑fuelled SMRs deployed near industrial clusters, supporting renewables while also supplying high-temperature heat for hydrogen production or district heating. Suggested refuelling intervals of 8–15 years would reduce the regular fuel-handling cadence familiar from large light-water reactors.

That future comes with its own vulnerabilities: more, smaller units mean more sites to protect, more movements of specialised fuel, and a strong dependency on a still-developing HALEU supply chain. Governments assessing HALEU-heavy pathways therefore have to weigh not only costs and carbon targets, but also long-term resilience and security of supply.